Camouflaging and autism: Conceptualisation and methodological issues

It has been suggested that there is poor clarity of the 'camouflaging' concept in autism research, and potential confounding of its measurement tools, such as the Camouflaging Autistic Traits Questionnaire (CAT-Q). A critical review of 389 studies was conducted to investigate these potential conceptual and methodological issues. The findings question whether the same construct has been investigated across studies, as there is inconsistency in: (a) which terms are used; (b) reference made to established conceptual literature; (c) how terms are used (e.g. interchangeably or distinctly); and (d) how terms are defined. Although the CAT-Q has excellent reliability, there is mixed support for its validity, which is evidenced by its confounding by other constructs (e.g. social anxiety) and its limited autism-specificity. The validity of informant discrepancy measures of camouflaging is also questioned due to insufficient reference to established methodology. Finally, the generalisability of camouflaging to the overall autistic population is unclear due to: (a) overrepresentation of autistic females diagnosed during adulthood; and (b) underrepresentation of autistic people with co-occurring intellectual or language difficulties, and those with greater support needs. These issues are considered both in terms of their clinical relevance and how future research might resolve them.Lay AbstractMany autistic people have reported using 'camouflaging' strategies to adapt or cope within the non-autistic social world and avoid being negatively judged by other people. However, many terms have been used synonymously with camouflaging, such as masking, compensation and impression management. Due to this confusion about which terms to use, there is some suggestion that there is poor clarity and understanding of the camouflaging concept, and that this may contribute to inaccuracy of the tools used to measure this behaviour. We review 389 previous studies to examine these concerns. Our findings confirm this lack of clarity by showing that studies are inconsistent in: (a) which terms they used to refer to behavioural strategies that resemble camouflaging; (b) whether they referred to existing literature; (c) whether they used different terms to refer to the same concept or to separate types of behaviour; and (d) how they defined the terms that they used. Our findings also question the accuracy of camouflaging measurement tools, as these tools may also be measuring other behaviours (e.g. social anxiety) that are not only experienced by autistic people. We also find that camouflaging studies have mostly focused on autistic females with no accompanying cognitive or language difficulties, and who have received their diagnosis in adulthood. Although camouflaging may contribute to the underdiagnosis of some autistic females, most autistic people are male and are diagnosed during early childhood, and a large number of autistic people do experience those other difficulties. Because of this, it is not clear whether the findings of previous camouflaging studies can apply to all autistic people. We provide some suggestions for how researchers might resolve some of these issues in their future studies, and we consider what the findings mean for clinicians who work with autistic people.